[core] Small optimizations from accelerated DAG experiment #41418
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Eric Liang <ekhliang@gmail.com>
Signed-off-by: Eric Liang <ekhliang@gmail.com>
rynewang
reviewed
Nov 27, 2023
rkooo567
reviewed
Nov 28, 2023
rkooo567
added
the
@author-action-required
rynewang
reviewed
Nov 28, 2023
|
Addressed. Good idea on the static assert, I was trying to get |
ericl
removed
the
@author-action-required
rkooo567
approved these changes
Nov 28, 2023
Signed-off-by: Eric Liang <ekhliang@gmail.com>
Signed-off-by: Eric Liang <ekhliang@gmail.com>
rkooo567
reviewed
Nov 29, 2023
| @@ -393,12 +393,12 @@ void ReferenceCounter::RemoveLocalReferenceInternal(const ObjectID &object_id, | |||
| RAY_CHECK(!object_id.IsNil()); | |||
| auto it = object_id_refs_.find(object_id); | |||
| if (it == object_id_refs_.end()) { | |||
| RAY_LOG(WARNING) << "Tried to decrease ref count for nonexistent object ID: " | |||
| << object_id; | |||
| RAY_LOG_EVERY_MS(WARNING, 5000) | |||
There was a problem hiding this comment.
Btw, I assume these warning logs are impacting the performance, but does that mean it is more frequently printed when an accelerated DAG is used?
There was a problem hiding this comment.
It's definitely an issue we need to look into in the future, once we fix the ref counting.
Signed-off-by: Eric Liang <ekhliang@gmail.com>
Signed-off-by: Eric Liang <ekhliang@gmail.com>
Signed-off-by: Eric Liang <ekhliang@gmail.com>
This reverts commit 53cfcbb.
ericl
added
the
tests-ok
pcmoritz
approved these changes
Nov 30, 2023
raulchen
approved these changes
Nov 30, 2023
fishbone
approved these changes
Nov 30, 2023
| os.environ.get("RAY_ENABLE_AUTO_CONNECT", "") != "0" | ||
| and not ray.is_initialized() | ||
| ): | ||
| if enable_auto_connect and not ray.is_initialized(): |
There was a problem hiding this comment.
+1
Beat me to fixing this. It's a small thing but notoriously showing up in every profile
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Why are these changes needed?
This ports a few micro-optimizations from the accelerated DAG experiment branch (https://github.com/ray-project/ray/pull/40991/files). The optimizations here help the most for accelerated DAG programs, but also help a bit to reduce get()/put() call overheads for normal Ray programs:
Before:
After: